Fix Oracle DR test ORA-00600 error by forcing service shutdown in cleanup

Problem: DR weekly test failed with ORA-00600 [kcbzib_kcrsds_1] when executed
via cron, but succeeded when run manually. Error occurred during "ALTER DATABASE
OPEN RESETLOGS" step after successful restore and recovery.

Root Cause Analysis:
- Manual test (12:09): Undo initialization = 0ms, no errors
- Cron test (10:45): Undo initialization = 2735ms, ORA-00600 crash
- Alert log showed: "Undo initialization recovery: err:600"
- Oracle instance was in inconsistent state from previous run

The cleanup_database.ps1 script had an "optimization" that preserved the
running Oracle service to "save ~30s startup time". This left the service
in an inconsistent state between test runs, causing Oracle to crash when
attempting to open the database with RESETLOGS.

Solution:
Modified cleanup_database.ps1 to ALWAYS stop Oracle service completely:
1. SHUTDOWN ABORT the instance (not just when /AFTER flag)
2. Stop-Service OracleServiceROA (force clean state)
3. Kill remaining oracle processes
4. Service starts fresh during restore (clean Undo initialization)

Changes:
- Removed if/else branch that skipped shutdown before restore
- Always perform full shutdown regardless of /AFTER parameter
- Updated messages to reflect clean state approach
- Added explanation: "This ensures no state inconsistencies (prevents ORA-00600)"

Testing: Manual test confirmed clean 0ms Undo initialization after fix.

Related: Works in conjunction with weekly-dr-test-proxmox.sh PATH fix (commit 34f91ba)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

oracle/standby-server-scripts/cleanup_database.ps1

@@ -40,56 +40,50 @@ New-Item -ItemType Directory -Path "D:\oracle\temp" -Force | Out-Null
 New-Item -ItemType Directory -Path "D:\oracle\logs" -Force | Out-Null
 
 Write-Host ""
-if ($afterRestore) {
-    Write-Host "[1/6] Shutting down database (cleanup AFTER restore)..."
+Write-Host "[1/6] Shutting down database and stopping service..."
+
+# Check if Oracle service exists
+$service = Get-Service -Name "OracleServiceROA" -ErrorAction SilentlyContinue
+if ($service) {
+    Write-Host "    Oracle service found, ensuring clean shutdown..."
     
-    # Check if Oracle service exists
-    $service = Get-Service -Name "OracleServiceROA" -ErrorAction SilentlyContinue
-    if ($service) {
-        Write-Host "    Oracle service found, attempting shutdown..."
-        
-        # Shutdown instance using SQL*Plus
-        $shutdownSQL = "WHENEVER SQLERROR CONTINUE`nSHUTDOWN ABORT;`nEXIT;"
-        try {
-            $shutdownSQL | & sqlplus -S / as sysdba 2>&1 | Out-Null
-            Start-Sleep -Seconds 2
-            Write-Host "    Instance shut down"
-        } catch {
-            Write-Host "    Shutdown command sent (errors ignored)"
-        }
-        
-        # Stop Oracle service to release file locks
-        if ($service.Status -eq "Running") {
-            Write-Host "    Stopping Oracle service to release file locks..."
-            try {
-                Stop-Service -Name "OracleServiceROA" -Force -ErrorAction Stop
-                Start-Sleep -Seconds 2
-                Write-Host "    Service stopped"
-            } catch {
-                Write-Host "    WARNING: Failed to stop service: $_" -ForegroundColor Yellow
-            }
-        }
-        
-        # Force kill any remaining Oracle processes
-        Get-Process -Name "sqlplus" -ErrorAction SilentlyContinue | Stop-Process -Force -ErrorAction SilentlyContinue
-        Get-Process -Name "oracle" -ErrorAction SilentlyContinue | Stop-Process -Force -ErrorAction SilentlyContinue
-    } else {
-        Write-Host "    Oracle service not found, skipping shutdown"
+    # Shutdown instance using SQL*Plus (always, not just /AFTER)
+    $shutdownSQL = "WHENEVER SQLERROR CONTINUE`nSHUTDOWN ABORT;`nEXIT;"
+    try {
+        $shutdownSQL | & sqlplus -S / as sysdba 2>&1 | Out-Null
+        Start-Sleep -Seconds 2
+        Write-Host "    Instance shut down (ABORT for fast cleanup)"
+    } catch {
+        Write-Host "    Shutdown command sent (errors ignored)"
     }
+    
+    # ALWAYS stop Oracle service to ensure clean state
+    if ($service.Status -eq "Running") {
+        Write-Host "    Stopping Oracle service to ensure clean state..."
+        try {
+            Stop-Service -Name "OracleServiceROA" -Force -ErrorAction Stop
+            Start-Sleep -Seconds 3
+            Write-Host "    Service stopped successfully"
+        } catch {
+            Write-Host "    WARNING: Failed to stop service: $_" -ForegroundColor Yellow
+        }
+    } else {
+        Write-Host "    Service already stopped"
+    }
+    
+    # Force kill any remaining Oracle processes to ensure clean state
+    Write-Host "    Cleaning up any remaining Oracle processes..."
+    Get-Process -Name "sqlplus" -ErrorAction SilentlyContinue | Stop-Process -Force -ErrorAction SilentlyContinue
+    Get-Process -Name "oracle" -ErrorAction SilentlyContinue | Stop-Process -Force -ErrorAction SilentlyContinue
     Start-Sleep -Seconds 2
+    Write-Host "    All Oracle processes terminated"
 } else {
-    Write-Host "[1/6] Skipping instance shutdown (cleanup BEFORE restore)"
-    Write-Host "    Instance and service left in current state"
-    Write-Host "    Restore script will handle service state properly"
+    Write-Host "    Oracle service not found, will be created during restore"
 }
 
-Write-Host "[2/6] Oracle service preserved for reuse"
-if ($afterRestore) {
-    Write-Host "    Service stopped to release file locks"
-} else {
-    Write-Host "    Service remains in current state (running or stopped)"
-    Write-Host "    Optimization: If running, restore saves ~30s startup time"
-}
+Write-Host "[2/6] Oracle service stopped (clean state for restore)"
+Write-Host "    Service will be started fresh during restore"
+Write-Host "    This ensures no state inconsistencies (prevents ORA-00600)"
 
 Write-Host "[3/6] Deleting database files + SPFILE..."
 Write-Host "    Deleting datafiles..."
@@ -124,16 +118,16 @@ Write-Host ""
 Write-Host "Current state:"
 Write-Host "  [YES] Oracle software installed"
 Write-Host "  [YES] PFILE exists (C:\Users\oracle\admin\ROA\pfile\initROA.ora)"
-Write-Host "  [YES] Oracle service (preserved, will be reused)"
+Write-Host "  [YES] Oracle service (STOPPED for clean restore)"
 Write-Host "  [NO]  SPFILE (deleted to ensure PFILE startup)"
 Write-Host "  [NO]  Database files (will be restored from backups)"
 Write-Host "  [NO]  Control files (will be restored from backups)"
 Write-Host "  [NO]  Datafiles (will be restored from backups)"
 Write-Host ""
-Write-Host "VM is now in CLEAN STATE (service preserved)!"
+Write-Host "VM is now in CLEAN STATE (service stopped, ready for fresh start)!"
 Write-Host ""
 Write-Host "Next step: Run D:\oracle\scripts\rman_restore_from_zero.ps1"
-Write-Host "  (It will reuse the existing Oracle service and restore the database)"
+Write-Host "  (It will start the service fresh and restore the database)"
 Write-Host ""
 
 exit 0